Extractive question answering task. Given a question and an associated paragraph systems should locate the shortest span of text that contains the answer. There are no unanswerable questions. The dataset contains texts extracted from the Spanish Wikipedia, encyclopedic articles, newswire articles from Wikinew and news.
Publication
Asier Gutiérrez Fandiño, Jordi Armengol-Estapé, Marc Pàmies, Joan Llop-Palao,Joaquín Silveira-Ocampo,Casimiro Pio Carrino, Carme Armentano-Oller, Carlos Rodriguez-Penagos, Aitor Gonzalez-Agirre, Marta Villegas (2022) Procesamiento del Lenguaje Natural, Revista nº 68, marzo de 2022, pp. 39-60.
Language
Spanish
NLP topic
Abstract task
Dataset
Year
2022
Publication link
Ranking metric
F1
Task results
| System | F1 |
|---|---|
| distillbert-base-spanish-uncased | 0.5300 |
| distilbert-base-multilingual-cased | 0.5500 |
| xlm-roberta-base | 0.6670 |
| ixambert-base-cased | 0.7065 |
| bert-base-multilingual-cased | 0.7080 |
| bert-base-spanish-wwm-cased | 0.7100 |
| PlanTL-GOB-ES-roberta-base-bne | 0.7360 |
| bertin-roberta-base-spanish | 0.7300 |
| PlanTL-GOB-ES-roberta-large-bne | 0.7670 |
| xlm-roberta-large | 0.7698 |
